We have a new blog entry up that is a non-technical summary of the strengths and weaknesses of Lightning Network. The agenda for starting work on LN is included in the current 1.0.0 release and stakeholders can vote on it. Feel free to post here if you have any feedback or comments.
Thanks for taking the time to write and publish the blog post about LN. I'm looking forward to its support in Decred.
Great post jy-p, this should certainly help to get peoples heads wrapped around LN. It is a complex topic and your post provides a great intro without being overwhelming. I can recommend this article for anybody wanting to go one level deeper into LN and still without being overwhelming: https://medium.com/@AudunGulbrands1/lightning-faq-67bd2b957d70 If anybody would like to learn the basics by way of analogy, there is a "Let's talk Bitcoin" episode which compares Lightning Network to opening a tab at a bar: https://letstalkbitcoin.com/blog/post/lets-talk-bitcoin-286-drinks-on-a-lightning-network (In my experience this is the best way to introduce non-techy people to LN)
For some reason when I've been thinking about lightning network before I kind of overlooked the cross chain atomic swaps. That might be the most interesting part to me. I guess I should go read the whitepaper more carefully
From http://lightning.network/ > Cross Blockchains. Cross-chain atomic swaps can occur off-chain instantly with heterogeneous blockchain consensus rules. So long as the chains can support the same cryptographic hash function, it is possible to make transactions across blockchains without trust in 3rd party custodians. Bitcoin uses SHA256 and Decred uses BLAKE-256. Can they still somehow participate in a swap?
Yes they will be able to work together. That's not referring to the hash function used for mining coins, rather the hash function used by the primitives that carry out the swap. Swaps do not need to use the same hash function as PoW.
is there actually any timeframe on this? I know it's being worked on - but are progress reports planned, or just one day a pump and then ninja launch as seems to be happening?
Answer from davecgh on the Reddit AMA session: "The short answer is, as the roadmap calls out, LN integration will likely be voted in and live before the end of the year. The slightly longer answer is that necessary code will probably land to master within 3-4 months and then there will need to be an on-chain vote to accept the underlying consensus changes or not which is roughly another 2 months (1 month for the vote itself, 1 month for the lockin period) That said, I do want to point out that when it comes to consensus changes, we are extremely careful and will not release them until they are carefully and painstakingly reviewed and tested because there is zero margin for error there. LN work involves consensus changes, so that work will not land until it's ready. I mention this in comparison to other areas such as websites and GUI glitches where mistakes, while not ideal, aren't super critical."
Yeah, timelines are always a little rough for consensus code since a delay is much better than a mistake there.